<div dir="ltr"><div dir="ltr">On Mon, May 20, 2019 at 11:42 AM D. Hugh Redelmeier via talk <<a href="mailto:talk@gtalug.org">talk@gtalug.org</a>> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">| From: Russell Reiter via talk <<a href="mailto:talk@gtalug.org" target="_blank">talk@gtalug.org</a>><br>
<br>
| Optane was intended to be a cache memory to increase the performance of<br>
| conventional spinning HD's under Windows OS. However, I've been booting<br>
| Fedora from both a 500gb SSD and 32gb Optane Nvme as I tinker with my own<br>
| desktop.<br>
<br>
I think Optane is just a brand name and that brand name gets attached<br>
to several different things, some of which are not yet being sold.<br></blockquote><div><br></div><div>True enough, sometimes it's hard to sort specifications from hyperbole. </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
The original promise was: non-volatile memory that would get speed<br>
close to RAM and price close to flash. (They also need better<br>
durability than flash.) There's a big gap between those, and anything<br>
in between ought to have a market. They just haven't been able to<br>
accomplish much.<br>
<br>
- they wanted to produce things that fit in RAM sockets. That meant<br>
either<br>
<br>
(1) no change from RAM interface (unlikely, but suggested-by-omission<br>
in early marketing) or<br>
<br>
(2) memory interfaces that were augmented to support the new protocols<br>
(delivered with some new Xeons, I think)<br>
<br>
- the NVMe stuff has had much worse durability than originally<br>
promised.<br>
<br>
- the NVMe stuff has had quite small capacity compared with most SSDs<br>
<br>
- NVMe SSD has mostly been fast enough that the Optane stuff isn't<br>
compellingly better<br>
<br>
| Most certainly booting to a login prompt is fractionally quicker<br>
| on the Nvme than on the conventional SSD.<br>
<br>
That's what I'd expect. But I'll admit to no experience with Optane<br>
and I haven't been following it closely.<br>
<br>
| However recently I up-sized my<br>
| Nvme and have populated my M.2 slots with a 250gb WD black Nvme for boot<br>
| and now added an additional 1TB to the second M.2 slot with F29 still on<br>
| the SSD. Copying a 100gb image to the 1TB drive really hit performance tho<br>
<br>
SSDs come with different performance trade-offs. Most inexpensive<br>
SSDs have (on-board) controllers with only small amounts of RAM. This<br>
makes them slow down a lot after a modest burst of intensive writing.<br>
That's a fine trade-off for many of us but not for all workloads.<br>
<br>
You can pay more and get SSDs with enough RAM to not have this<br>
performance problem. I don't know enough to give specific advice.<br>
<br>
I have recently learned that some cheap NVMe drives can ask the OS to<br>
allocate system RAM for the exclusive use of the controller. This<br>
isn't a conventional data cache, but something much stranger. It's<br>
called "Host Memory Buffer" (HMB). I think that vendors don't explain<br>
it because they think consumers won't understand it.<br>
<br>
HMB might be a great trade-off, or a horrible hack. I don't know.<br>
What happens when the power fails? Or when the system crashes /<br>
reboots?<br>
<br>
Recent Linux and Windows support HMB. I assume that UEFI firmware does<br>
not. So use of HMB must be an optional speed-up.<br>
<br>
| and that was probably due to the lack of a decent heat sink.<br>
| I just ordered a hteatsink fro the internal 1TB<br>
<br>
Why do you think that this was heat-related? It might be, but that<br>
would not be my first guess. (I am not an expert on this.)<br></blockquote><div><br></div><div>Apparently, at least on Crucial products, there is built in thermal monitoring which will throttle speeds. </div><div><br></div><div>Here's a link to the 1TB Nvme. On sale for the next few days for $147.00, about $40.00 less than I paid.</div><div> </div><div><a href="https://m.newegg.ca/crucial-p1-1tb/p/N82E16820156199?item=N82E16820156199&m_ver=1" target="_blank">https://m.newegg.ca/crucial-p1-1tb/p/N82E16820156199?item=N82E16820156199&m_ver=1</a></div><div><br></div><div>Here's a link to a review of Crucial's Thermal Throttling capacities<br><div><br></div><div><a href="https://www.techpowerup.com/reviews/Crucial/P1_NVMe_M.2_SSD_1_TB/7.html">https://www.techpowerup.com/reviews/Crucial/P1_NVMe_M.2_SSD_1_TB/7.html</a> </div></div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
| Perhaps with the NUC form factor heat might<br>
| be a problem on a larger sized Nvme but with USB-C you have wiggle room for<br>
| adaptation.<br>
<br>
The NUC form factor certainly reduces the heat disposal and thus<br>
limits components. But the main such component is the CPU. I've not<br>
heard of it being a problem for consumer 3.5" SSDs (what Evan intends<br>
to use) or NVMe drives.<br>
---<br>
Talk Mailing List<br>
<a href="mailto:talk@gtalug.org" target="_blank">talk@gtalug.org</a><br>
<a href="https://gtalug.org/mailman/listinfo/talk" rel="noreferrer" target="_blank">https://gtalug.org/mailman/listinfo/talk</a><br>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>Russell<br></div></div></div></div></div></div>